Harness acquires ChaosNative to bring 'chaos engineering' to DevOps workflows

The software delivery platform startup Harness Inc. today announced it has acquired ChaosNative Inc. to extend its scalable chaos engineering capabilities for enterprise development and operations teams.

Chaos engineering is the principle of approaching software production by building it to withstand unexpected or “chaotic” conditions, such as failures or disasters. In layman’s terms, this is called resiliency.

ChaosNative is the originator of the popular Cloud Native Computing Foundation’s open-source project LitmusChaos and the ChaosNative Litmus enterprise platform. Both of these platforms enable engineering teams to identify weaknesses in software infrastructure by introducing failures, resource outages, latency spikes and other chaos in a controlled way to test the outcomes.

Harness allows DevOps teams, which integrate development and information technology operations, to work in continuous delivery and integration, or CI/CD, workflows to speed up software deployment processes.

“At Harness, it has been our vision to expand our platform through more of the software CI/CD pipeline,” Jyoti Bansal, chief executive of Harness, told SiliconANGLE in an interview. “As we have been expanding into a broad platform, chaos engineering is a natural next step. It is a very popular practice, especially with microservices, which makes it very difficult to tell where things might break.”

Bansal explained that as companies rely more heavily on software to automate more things and begin to deploy more often the core use case of CI/CD is high velocity. In some cases, businesses may be deploying changes to their software 10 or 100 times a day. However, a lack of confidence in changes to the code could easily hinder that velocity.

“Velocity doesn’t happen if you don’t have confidence,” Basal said. “If you don’t have the confidence that things won’t break, then velocity slows down automatically. Chaos testing is very good at providing confidence. If you’ve made a change and all your chaos tests are running on your code, you can be reasonably sure that once it’s deployed it’s not going to break.”

The addition of ChaosNative’s capabilities to the Harness platform means that chaos testing done early in the workflow can provide higher confidence that deployed code will be resilient in production.

“All along, we believed that there is a lot to be done to make chaos engineering commonplace in DevOps. We kept saying – it is not just for Ops,” Uma Mukkara, co-founder of ChaosNative, wrote in the announcement. “Litmus has seen tremendous growth in the past year mainly in its use in CI/CD pipelines.”

When it came to choosing ChaosNative for acquisition, Bansal said one deciding factor was its cloud app capabilities and the company’s focus on Kubernetes microservices and clusters. Many resiliency issues happen in that layer, he explained, so getting them uncovered quickly is a must. Another reason is that the company’s products are designed for extreme ease-of-use for DevOps teams and there is also a very large community, it is one of the biggest in the industry, which means there’s a large go-to knowledge base and experts already present.

ChaosNative’s chaos engineering capabilities will become generally available to customers this quarter. Harness is integrating ChaosNative’s Litmus product directly into the software delivery platform and it will become part of the CI/CD pipeline to become a portion of the solution in order to ensure reliability and resilience going forward.